What is Semantic Operations?

Every organization operates by converting work into outcomes like revenue, growth, and product. Successful conversion depends on execution by teams and individuals and coordination through goals and leadership. Execution and coordination are constantly evaluated and adjusted based on measured comparison of expectations to reality. This is a tidy and vastly oversimplified version of "work", especially when you consider the number of documents, metrics, people, decisions, KPIs, processes, policies, and initiatives involved. But what if some of those things aren't quite clear? Their meaning isn't well-defined, so no one really understands them. Or alignment isn't reached, so the meaning varies depending on who you talk to. Now introduce complex data systems and AI — do things get better? Only if meaning is managed deliberately and explicitly.

Semantic Operations (SemOps) is a practical framework for businesses to align their technology systems, data, and organization so that people and machines work from accurate and consistent meaning. "Semantic" means meaning made explicit — concepts, rules, and relationships defined clearly enough for both people and machines to work with reliably.

The Problem SemOps Addresses

SemOps starts from a few observations. AI is fundamentally a form of analytics, and it performs exceptionally well where structures are coherent — code, data systems, well-defined domains. Organizations need similar coherence where most work is done and decisions are made, but typically don't have it. And now AI is expanding the scope of what counts as "data" — documents, conversations, decisions, code — on top of data systems that are already complex and difficult to manage.

The conditions that make AI work well are the same conditions that make organizations work well, with or without AI. SemOps targets that overlap. The goal is "better" — better decisions, higher quality, clearer alignment — and velocity, efficiency, and automation follow as byproducts. Achieving the right conditions can be difficult, but part of the playbook includes using AI to do this as well. The benefits start from day one.

Why SemOps? explores the full problem space and how I arrived at these conclusions.

A Framework

SemOps is a framework — a structured approach with methods, techniques, and principles that provide a path for organizations to benefit from data and AI.

It is not a platform, a SaaS product or a packaged solution. SemOps requires a shift in thinking, beyond technology, and including technical and non-technical roles. Organizations should evaluate their operations through a neutral understanding of capabilities rather than any particular tool, or branded product. Bundled systems and "enterprise" solutions are antithetical to the effective application of AI and agents. Technical systems should align to the specific business domain through transparent and simplified components, not generic platforms that impose their own structure.

SemOps isn't just guidance. It is concrete, technical solutions — architectural patterns, data governance methods, schema, AI agents, and implementation code — all available through SemOps-ai on GitHub. In fact, the framework seeks to make concrete much of what isn't now: business logic encoded into systems, meaning made measurable, and domain structure that both humans and machines can work with directly.

The Semantic Operations Framework

The framework is built on a mental model and three pillars. Strategic Data and Explicit Architecture create the conditions — coherent data systems, domain-aligned architecture, and transparent, governed systems. Semantic Optimization is where those conditions pay off: AI operates against a coherent foundation to measure, improve, and compound growth across the organization.

The Semantic Funnel

The Semantic Funnel is the mental model that simplifies and explains the framework as well as AI and analytics use-cases in general. It reduces complex data and AI scenarios to three entities and one knowledge process. It provides a shared vocabulary for understanding what organizations actually do with information, where meaning gets constructed, and where human judgment versus automation is most effective.

Strategic Data

Strategic Data addresses how organizations think about and manage data as a strategic asset. Data systems are inherently complex and cut across organizational boundaries — roles, teams, and priorities that don't naturally align. Strategic Data provides techniques for building shared understanding of data across those boundaries, and applies AI through the framework to make data governance and management easier and more effective.

Explicit Architecture

Explicit Architecture addresses how organizations encode their business structure into systems. When architecture explicitly reflects the business domain — its boundaries, relationships, and rules — technology choices and infrastructure become last-mile decisions rather than foundational ones. AI accelerates the work of encoding domain structure, and that structure makes AI more effective.

Semantic Optimization

Semantic Optimization is where AI does what it does best — measure, analyze, and act on structured information — while governance ensures agents operate with appropriate autonomy. The result is that people spend less time managing systems and more time on strategy, experimentation, and exploration.